code-reviewer

Review code diffs and pull requests for bugs and security issues.

1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/wnz99/claude-skills --skill code-reviewer-wnz99
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-reviewer
Source: https://github.com/wnz99/claude-skills/tree/main/skills/code-reviewer
Command: npx skills add https://github.com/wnz99/claude-skills --skill code-reviewer-wnz99

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps agents review code changes with a findings-first, evidence-backed approach so real bugs, regressions, and security issues are caught before they ship.

Core Features & Use Cases

  • Local change review: Inspect staged and unstaged diffs for correctness, maintainability, readability, and test gaps.
  • Remote PR review: Review pull requests by number or URL, then assess the full change set, context, and discussion history.
  • Looped validation workflows: Support repeated review cycles with verification commands, severity-ranked findings, and cross-model sub-agent feedback.
  • Use case: A developer asks for a review of a feature branch and needs concrete, line-referenced feedback on bugs, missing tests, and risky behavioral changes.

Quick Start

Ask the code-reviewer skill to review my current changes or a specific pull request and return only the highest-confidence findings first.

Frequently Asked Questions about code-reviewer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review code changes for bugs and security issues before a pull request?

Review code changes by inspecting staged and unstaged diffs to identify bugs, regressions, and security issues with severity-ranked feedback. This process catches maintainability problems and provides line-referenced findings before merging.

Can I review a remote pull request by its URL or number?

Yes, you can review remote pull requests by providing the specific URL or number. The review assesses the full change set, context, and discussion history to return evidence-backed findings and severity-ranked feedback.

How does cross-model validation work for debugging regressions?

Cross-model validation supports repeated review cycles by utilizing sub-agent feedback to verify findings. This looped workflow applies terminal-aware commands to ensure evidence-backed debugging of regressions and behavioral changes.

What is the best way to find missing tests and risky behavioral changes in a feature branch?

The best way to find missing tests is to run a focused review on the feature branch diff. This yields concrete, line-referenced feedback on bugs, missing tests, and risky behavioral changes with high confidence.

Does this code review process require specific dependencies or environments?

This code review process requires no explicit dependencies but relies on terminal-aware command selection and project-instruction discovery. It operates within your local environment to verify staged, unstaged, or remote diffs.

When should I not use automated code review for my pull request?

You should avoid automated code review when your changes lack a clear diff to inspect, such as binary files or unresolved merge conflicts. The process requires readable code diffs to generate evidence-backed, file and line-referenced findings.